0429 Sleep Misperception in Insomnia Phenotypes Based on Objective Sleep Duration in Young Adults

Abstract Introduction Prior research has suggested that the underestimation of total sleep time (TST), i.e., so-called sleep misperception, may be a trait feature of those with insomnia. Prior studies have also shown differences in the degree of sleep misperception across insomnia phenotypes, however, these findings have not been replicated in population-based samples of young adults. Methods We studied 270 young adults (median 25 years, 53% female, 24% racial/ethnic minority) from the Penn State Child Cohort who underwent a 9-hour polysomnography (PSG) recording, clinical history, and self-report surveys. Insomnia symptoms were defined as reports of difficulties initiating or maintaining sleep, insomnia diagnosis or complaint, and/or sleep medication use. PSG-measured short sleep duration was defined by the median of the sample (i.e., < 7-h), identifying normal sleep duration (NSD), short sleep duration (SSD), insomnia with normal sleep duration (INSD) and insomnia with short sleep duration (ISSD). Subjective TST was ascertained from a morning questionnaire completed immediately upon awakening from the PSG as well as from a retrospective survey of habitual sleep at home. A general linear model tested mean differences in TST-discrepancy across the four groups, while adjusting for sex, race/ethnicity, age, waist circumference, sleep apnea, cardiometabolic disorders, medication and substance use. Results INSD consistently underestimated TST, regardless of whether it was assessed upon awakening in the lab (-37.9±7.7 min; p< 0.001) or by reports of habitual sleep (-41.5±9.1 min; p< 0.001), while ISSD showed relative accuracy when TST was assessed upon awakening in the lab (8.9±7.8 minutes) and clear overestimation (85.9±9.2 minutes) when assessed by reports of habitual sleep. Conclusion Sleep misperception is a trait feature of individuals with INSD, who underestimate their sleep duration regardless of method of measurement used. These data suggest that the INSD phenotype may respond better to therapies addressing the cognitive and cortical underpinnings of this phenomenon. Support (if any) NIH (R01 HL136587, R01 MH118308, UL1 TR00127)